Model comparison
Claude Opus 5.5 vs GPT-6 Sol: why caching widens the gap
Claude Opus 5.5 and GPT-6 Sol launched the same day. Opus 5.5 lists at 2x Sol's prices, and its 1-hour cache writes stretch a coding session to 2.1x.
· Prices as of September 28, 2026
Claude Opus 5.5
Anthropic · Released September 22, 2026
Anthropic's recommended starting model for most work, built for long-running agentic coding. Anthropic says it matches Claude Fable 5.1 on most work at a much lower price.
Claude Opus 5.5 facts and comparisonsGPT-6 Sol
OpenAI · Released September 22, 2026
The mid-priced GPT-6 model, which OpenAI pitches for complex coding and agent workflows and which the Codex docs recommend for complex coding.
GPT-6 Sol facts and comparisons
The short answer
GPT-6 Sol is the cheaper model at half Claude Opus 5.5's input, output, and cache-write prices, and the example agentic coding session costs $2.10 on Sol against $4.40 on Opus 5.5, a 2.1x gap. Both charge $0.20 per million for a cache hit, yet the session gap grows past 2x because Anthropic's 1-hour cache writes carry a heavier premium. Opus 5.5 is the natural pick in Claude Code, where it is the default, and Sol in Codex, whose docs recommend it for complex coding.
Choose Claude Opus 5.5 if
- You work in Claude Code, where Opus 5.5 is the default model on Pro, Max, Team, and Enterprise plans.
- Your work is codebase-wide migrations or audits, which Anthropic names as a strength of Opus 5.5.
- Your prompts pass 272K input tokens, where GPT-6 Sol's rates rise and Opus 5.5 keeps standard rates up to 1M.
- You want fast mode as an option, a research preview on the Claude API at $8 input and $40 output.
Choose GPT-6 Sol if
- You want half the price on input, output, and cache writes: $2, $10, and $2.50 per million.
- You work in Codex, whose docs recommend GPT-6 Sol for complex coding and suggest it in place of GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.4.
- You mark cache breakpoints yourself and want every write at 1.25x input rather than Anthropic's 2x for a 1-hour write.
Side by side
Specs and prices
| Fact | Claude Opus 5.5 | GPT-6 Sol |
|---|---|---|
| Maker | Anthropic | OpenAI |
| API model id | claude-opus-5-5 | gpt-6-sol |
| Released | September 22, 2026 | September 22, 2026 |
| Status | Current | Current |
| Context window | 1M tokens | 1.05M tokens |
| Max output | 128K tokens | 128K tokens |
| Open weights | No | No |
| Input, per 1M tokens | $4 | $2 |
| Cache hit, per 1M | $0.20 | $0.20 |
| Cache write, per 1M | $5 (5-minute), $8 (1-hour) | $2.50 |
| Output, per 1M tokens | $20 | $10 |
| Runs in | Claude Code, Cursor, OpenCode, OpenRouter, and GitHub Copilot | Codex, OpenCode, OpenRouter, and GitHub Copilot |
Standard API rates in US dollars, as published by each maker (Claude Opus 5.5: September 26, 2026; GPT-6 Sol: September 28, 2026). Batch and priority tiers, taxes, and subscription plans are not included. Claude Opus 5.5: The full 1M context window is billed at standard rates. Fast mode, a research preview on the Claude API, costs $8 input and $40 output per million tokens. GPT-6 Sol: Requests over 272K input tokens cost 2x for input and cache and 1.5x for output, for the whole request.
Cost
What the same work costs
The same token counts, priced at each model’s published rates. Cached tokens are billed at each maker’s cache prices, so the session shows what caching is worth on each model.
Real sessions differ: the two models count the same code as different numbers of tokens, and reasoning settings change how much each one writes. Your own history is the real test.
| Workload | Claude Opus 5.5 | GPT-6 Sol |
|---|---|---|
| Agentic coding session, 100K input, 400K written to cache, 2M read from cache, 50K output | $4.40 | $2.10 |
| Large one-off review, 150K input with no cache hits, 10K output | $0.80 | $0.40 |
| Output-heavy generation, 30K input, 80K output | $1.72 | $0.86 |
| A month of sessions, 110 sessions: 5 a day, 22 working days | $484.00 | $231.00 |
| Where the session’s cost goes | ||
| Cache writes | $2.60 | $1.00 |
| Cache reads | $0.40 | $0.40 |
| Uncached input | $0.40 | $0.20 |
| Output | $1.00 | $0.50 |
- caching saves on the session with Claude Opus 5.5 (60%)
- $6.60
- caching saves on the session with GPT-6 Sol (62%)
- $3.40
Same launch day, same cache hit price
Claude Opus 5.5 and GPT-6 Sol were both released on September 22, 2026, and they share one price: $0.20 per million cached tokens. Anthropic reaches that figure by pricing an Opus 5.5 hit at 0.05x input, while OpenAI prices a Sol hit at the usual 0.1x of a lower input price.
Everything else on Opus 5.5 costs twice as much: $4 against $2 for input, $20 against $10 for output, and $5 against $2.50 for a standard cache write. The uncached review, $0.80 against $0.40, and output-heavy generation, $1.72 against $0.86, both show exactly 2x.
Why caching widens the gap to 2.1x
The example session reads 2M tokens from cache, which costs $0.40 on either model. With reads equal, you might expect the session gap to shrink below 2x. It grows to 2.1x instead, $4.40 against $2.10, and the writes are the reason.
Half of the session's 400K written tokens go to Anthropic's 1-hour cache at 2x input, $8 per million on Opus 5.5. OpenAI bills every Sol write at 1.25x, $2.50. Cache writes come to $2.60 on Opus 5.5 and $1.00 on Sol, so Opus 5.5 pays $1.60 more for writes, more than double Sol's figure.
Which cache Claude Code uses depends on how you sign in: the main conversation uses the 1-hour cache on a Claude subscription and the 5-minute cache with an API key. On the 5-minute cache, Opus 5.5 writes at 1.25x input, exactly twice Sol's write price, and the equal reads bring the session gap under 2x. Over 110 sessions a month the example comes to $484.00 on Opus 5.5 and $231.00 on Sol, a $253.00 difference.
How Anthropic and OpenAI position these two models
Anthropic calls Opus 5.5 its recommended starting model for most work and says "It performs at the level of Claude Fable 5.1 on most work and costs 40% less to run than Opus 5." It lists agentic coding, computer use, and knowledge work as strengths. OpenAI says GPT-6 Sol is "Built to power complex coding and agentic workflows" and prices it between GPT-6 Astra and GPT-6 Luna.
Both default to medium effort, Sol in both the API and Codex, and Anthropic notes that changing Opus 5.5's effort keeps the prompt cache. Context windows are similar, 1M on Opus 5.5 and 1.05M on Sol, which accepts up to 922K input tokens, and both write up to 128K. Opus 5.5's tokenizer counts about 30% more tokens than earlier Claude models for the same text, so one prompt produces different token counts on each.
Prompt caching
How each maker bills cached tokens
Anthropic
Claude caches a prompt prefix up to a breakpoint. One top-level cache_control field places the breakpoint automatically and moves it as the conversation grows, or you can mark up to 4 blocks yourself. Claude Code manages caching for you.
A 5-minute cache write costs 1.25x the input price and a 1-hour write costs 2x. A cache hit costs 0.1x input on most models, 0.05x on Claude Opus 5.5, and 0.025x on Claude Fable 5.1 and Claude Mythos 5.1. Every hit restarts the cache lifetime at no charge.
Anthropic's rule of thumb: a 5-minute write pays for itself after one cache read, and a 1-hour write after two.
In Claude Code, the main conversation uses the 1-hour cache on a Claude subscription and the 5-minute cache with an API key. Each model has its own cache, so switching models starts over.
Source: Anthropic: Prompt caching
OpenAI
Prompt caching is on by default. From GPT-5.6 on you can also mark up to four explicit cache breakpoints, while GPT-5.5 and earlier cache automatically only.
From GPT-5.6 on, a cache write costs 1.25x the uncached input price and a cache hit costs 0.1x. GPT-5.5 and earlier add no charge for writing the cache: written tokens are billed as ordinary input, and a hit costs 0.1x on the models compared here.
On GPT-5.6 and later, a cached prefix stays reusable for at least 30 minutes after its last use, and caching starts at 1,024 input tokens.
Source: OpenAI: Prompt caching
Your own numbers
See what Claude Opus 5.5 and GPT-6 Sol really cost you.
everyaitoken reads your Claude Code, Cursor, OpenCode, OpenRouter, and Codex history on your Mac and prices every request at API rates, with what caching saved or cost. $9 once.
FAQ
Questions
Is GPT-6 Sol half the price of Claude Opus 5.5?
Per token, yes, for input, output, and cache writes. Cache hits cost $0.20 per million on both. Because Anthropic's 1-hour writes cost more, the example agentic session comes out at 2.1x, $4.40 against $2.10.
Were Claude Opus 5.5 and GPT-6 Sol released on the same day?
Yes, both on September 22, 2026. Opus 5.5 is now Claude Code's default model, and Codex suggests moving to GPT-6 Sol from GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.4.
Where can I use each model?
Both are offered in OpenCode, OpenRouter, and GitHub Copilot. Beyond those, Opus 5.5 is in Claude Code and Cursor, and GPT-6 Sol is in Codex. Switching between them mid-task starts a new cache, since each model keeps its own.
How do I see what caching saves me on each?
The example session saves $6.60 on Opus 5.5 and $3.40 on Sol compared with paying full input price for every token. For your own numbers, EveryToken reads Claude Code and Codex history on your Mac and shows, per model, what caching saved or cost at API rates. A subscription plan charges differently from these API-equivalent figures.
Sources
- Anthropic: Pricing
- Anthropic docs: Claude Opus 5.5
- Anthropic: Claude Opus 5.5
- Anthropic docs: Fast mode
- Claude Code docs: Model configuration
- Cursor docs: Claude Opus 5.5
- OpenRouter: Claude Opus 5.5
- OpenCode docs: Zen
- GitHub Docs: Supported AI models in Copilot
- OpenAI: API pricing
- OpenAI docs: GPT-6 Sol
- OpenAI: Introducing GPT-6 Sol and Luna
- OpenAI: API changelog
- Codex docs: Models
- OpenRouter: GPT-6 Sol
- Anthropic: Prompt caching
- OpenAI: Prompt caching